The objective of the proposed study is to test whether the efficacy of locomotor training will be improved through the application of constraint induced forced use of the affected leg during locomotor training. We will also identify the neural mechanisms underlying the improvements in locomotor function after training in individuals post-stroke. Locomotor training using a treadmill is a promising technique that provides a safe and convenient environment for improving walking capability in individuals post-stroke. While the improvements in walking function after treadmill training are statistically significant, the functional gains are relatively small for many patients. One of primary reasons of the less effectiveness of treadmill training may be due to the compensatory motor strategies employed by patients during locomotor training, i.e., patients with hemiparesis often rely more on the unaffected leg for performing bipedal walking during treadmill training. Repetitive practice in thi manner may actually lead to reinforcing the compensatory motor strategies, which results in limited improvement in motor control of the affected leg, resulting in limited functional gains aftr training, which suggests a need of developing new training paradigms in order to maximize functional gains. Constraint induced movement therapy (CIMT) has been utilized to improve motor function of the affected arm in individuals post-stroke through forced use of their affected arm and by restricting movements of the unaffected arm. Previous studies have shown promising improvements in motor skills and in the use of the affected arm and hand in daily activities after CIMT. However, such interventions have not been effectively applied to lower limb training in individuals post-stroke due to the strong coupling between the two legs during bipedal gait and the risk of falling. Thus, we propose to develop a novel strategy to test CIMT for lower limb training in individuals post-stroke. Specifically, we will apply a controlled force to te pelvis and/or unaffected leg, which will serve to overcome the compensatory motor strategies employed by patients, and induce forced use of the affected leg during locomotor training. Our central hypothesis is that the efficacy of locomotor training will be improved by applying a controlled assistance force to the pelvis and/or resistance force to the unaffected leg to reduce the compensatory movements of the unaffected leg, and induce forced use of the affected leg of individuals post-stroke during training. Results from this study will lead to an innovative clinica therapy paradigm aimed at improving locomotor function in individuals post-stroke. We expect that this study will demonstrate the feasibility of a novel robotic training strategy, i.e., applyig a controlled force to the pelvis and/or the unaffected leg to induce forced use of the affected leg during locomotor training. The successful completion of the proposed study will have a high potential to make a significant impact on the field of locomotor rehabilitation in individuals post stroke through the application of a novel treatment paradigm. This paradigm may also be applied to other patient populations, such as patients with hemiparetic cerebral palsy.